BTIS-NET, SERESU-NET, and AGRESU-NET: A Comparative Study of Deep Learning Architectures for Brain Tumor Segmentation
Keywords:
Brain tumor segmentation, U-Net, convolutional neural network, AGRESU-NET, SERESU-NET, BTIS-NET, MRI, BraTS dataset, deep learning, attention mechanismAbstract
One of the most important steps in clinical diagnosis and treatment planning is the automated and precise segmentation of brain tumors using magnetic resonance imaging (MRI). Deep learning-based solutions are required since manual segmentation is laborious and subject to inter-observer variability. Three sophisticated convolutional neural network designs for brain tumor segmentation are compared in this paper: AGRESU-NET (Attention-Gated Residual U-Net), SERESU-NET (Squeeze-and-Excitation Residual U-Net), and BTIS-NET (Brain Tumor Inception Segmentation Network). The BraTS 2021 dataset, which consists of multi-modal MRI images with tumor subregion annotations, was used to assess the models. Sensitivity, Hausdorff Distance (HD), and Dice Similarity Coefficient (DSC) were used to evaluate performance. With an average DSC of 0.91 for total tumor segmentation and decreased HD, the experimental findings showed that AGRESU-NET performed better than the other models, demonstrating the usefulness of residual encoding and attention processes in segmenting complicated tumor geometries. The paper emphasizes how architectural developments have improved automated brain tumor segmentation systems' accuracy and dependability
Downloads
References
J. Long, E. Shelhamer, and T. Darrell, “Fully convolutional networks for semantic segmentation,” in Proc. IEEE CVPR, pp. 3431–3440, 2015.
O. Ronneberger, P. Fischer, and T. Brox, “U-net: Convolutional networks for biomedical image segmentation,” in Medical Image Computing and Computer-Assisted Intervention – MICCAI 2015, Springer, pp. 234–241, 2015.
G. Litjens et al., “A survey on deep learning in medical image analysis,” Med. Image Anal., vol. 42, pp. 60–88, Dec. 2017.
B. H. Menze et al., “The multimodal brain tumor image segmentation benchmark (BRATS),” IEEE Trans. Med. Imaging, vol. 34, no. 10, pp. 1993–2024, Oct. 2015.
F. Isensee et al., “nnu-net: Self-adapting framework for U-Net-based medical image segmentation,” Nature Methods, vol. 18, pp. 203–211, 2021.
C. Szegedy et al., “Going deeper with convolutions,” in Proc. IEEE CVPR, pp. 1–9, 2015.
J. Hu, L. Shen, and G. Sun, “Squeeze-and-excitation networks,” in Proc. IEEE CVPR, pp. 7132–7141, 2018.
K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,” in Proc. IEEE CVPR, pp. 770–778, 2016.
O. Oktay et al., “Attention U-Net: Learning where to look for the pancreas,” arXiv preprint, arXiv:1804.03999, 2018.
Y. Yan, W. Ding, and Y. Li, “SEResU-Net for multimodal brain tumor segmentation,” in Proc. ISBI, pp. 672–675, 2021.
X. Wang et al., “Residual attention network for image classification,” in Proc. IEEE CVPR, pp. 3156–3164, 2017.
S. Roy et al., “QuickNAT: A fully convolutional network for quick and accurate segmentation of neuroanatomy,” NeuroImage, vol. 186, pp. 713–727, 2019.
N. H. Tajbakhsh et al., “Convolutional neural networks for medical image analysis: Full training or fine tuning,” IEEE Trans. Med. Imaging, vol. 35, no. 5, pp. 1299–1312, May 2016.
D. Nie et al., “Medical image synthesis with context-aware generative adversarial networks,” in Medical Image Computing and Computer-Assisted Intervention – MICCAI 2017, Springer, pp. 417–425.
Vaswani et al., “Attention is all you need,” in Advances in Neural Information Processing Systems, vol. 30, 2017.
J. Cheng et al., “Enhanced performance of brain tumor segmentation via attention residual learning,” Neurocomputing, vol. 405, pp. 152–162, 2020.
D. Zhou et al., “Multi-class brain tumor segmentation using 3D residual attention U-Net,” IEEE Access, vol. 8, pp. 58533–58541, 2020.
S. Bakas et al., “Advancing the cancer genome atlas glioma MRI collections with expert segmentation labels and radiomic features,” Scientific Data, vol. 4, 170117, 2017.
X. Liu et al., “MS-Net: Multi-site network for improving brain tumor segmentation on multi-site MRI,” IEEE J. Biomed. Health Inform., vol. 26, no. 2, pp. 681–691, 2022.
D. Karimi, G. Salcudean, “Reducing the Hausdorff distance in medical image segmentation with convolutional neural networks,” IEEE Trans. Med. Imaging, vol. 39, no. 2, pp. 499–513, Feb. 2020.
M. Havaei et al., “Brain tumor segmentation with deep neural networks,” Med. Image Anal., vol. 35, pp. 18–31, Jan. 2017.
B. Zhao et al., “Exploring domain shift in segmentation of brain tumors with deep neural networks,” Front. Neurosci., vol. 13, pp. 584, 2019.
Z. Jiang et al., “Two-stage cascaded U-Net: 1st place solution to BraTS challenge 2019,” arXiv preprint, arXiv:1910.08056, 2019.
Myronenko, “3D MRI brain tumor segmentation using autoencoder regularization,” in Proc. MICCAI BraTS Workshop, 2018.
K. McKinley et al., “Ensembles of densely-connected CNNs with label-uncertainty for brain tumor segmentation,” in Proc. MICCAI BraTS Workshop, 2019.
S. Rezaei et al., “Multi-path U-Net with dense blocks for brain tumor segmentation,” Brain Informatics, vol. 9, no. 1, pp. 1–11, 2022.
Downloads
Published
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution 4.0 International License.
You are free to:
- Share — copy and redistribute the material in any medium or format
- Adapt — remix, transform, and build upon the material for any purpose, even commercially.
Terms:
- Attribution — You must give appropriate credit, provide a link to the license, and indicate if changes were made. You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use.
- No additional restrictions — You may not apply legal terms or technological measures that legally restrict others from doing anything the license permits.
